AI-powered LED system provides stable wireless power to indoor IoT devices

The world’s first self-adaptive dual-mode light-emitting diode (LED)-based optical wireless power transfer system that operates seamlessly in both dark and bright lighting conditions has been developed by scientists at Science Tokyo. Together with artificial intelligence-powered image recognition, the system can efficiently power multiple devices in sequence without interruption. Being LED-based, it provides a low-cost and secure solution ideal for building sustainable indoor Internet of Things infrastructures.

With the rapid development of the Internet of Things (IoT), the demand for efficient and flexible power solutions is also increasing. Traditional power delivery methods such as batteries and cable connections have many drawbacks. Batteries require frequent charging and replacement, and cables limit device mobility.

Optical wireless power transfer (OWPT) is a new technology that can address these limitations. In OWPT, energy is transmitted through free space without the use of physical wires by converting electricity to light for transmission and using a photovoltaic (PV) receiver to convert the light back into electricity.

Most of the current OWPT research focuses on laser-based systems. However, for indoor IoT scenarios, laser systems are not suitable unless special safety technology is developed, as OWPT systems must comply with strict maximum permissible exposure regulations to prevent eye and skin hazards.

In contrast, light-emitting diode (LED)-based OWPT systems are inherently safe, provide reliable power transfer, are easy to control, cost-effective, and have a long lifespan. However, these systems suffer from power loss over long distances and inconsistent performance under changing ambient lighting conditions.

To overcome these limitations, Professor Tomoyuki Miyamoto and postdoctoral researcher Mingzhi Zhao of the Future Science and Technology Fusion Laboratory at Tokyo University of Science (Science Tokyo) developed an innovative LED-based OWPT system. “We designed a dual-mode adaptive OWPT system that automatically adapts to both bright and dark indoor environments while enabling safe and efficient power delivery to multiple IoT devices,” explains Miyamoto. Their research was published in the journal Optics Express on October 24, 2025.

To overcome power loss during long-distance transmission, the proposed system utilizes an adaptive lens system with a double-layer lens configuration consisting of a liquid lens and an imaging lens with adjustable focal length. This setting automatically adjusts the beam spot size based on receiver distance and size to ensure optimal power transfer.

To precisely aim the light beam, the system utilizes an adjustable reflector that can be rotated independently in horizontal and vertical directions using two series-connected stepper motors. To precisely align towards the photovoltaic (PV) receiver, the researchers employed a depth camera with both RGB and infrared (IR) sensors. The RGB sensor detects the position of the PV receiver, and the IR sensor identifies the illuminated spot of the beam. This allows the control system to orient the reflector toward the target receiver.

To ensure continuous operation in both bright and dark conditions, the PV receiver is equipped with a retroreflective (RF) sheet around its edges. These sheets reflect the IR light emitted by the depth camera’s IR projector and create a clear outline of each PV receiver. This allows the system to accurately detect the shape and location of the receiver, isolate the target PV area, and minimize interference from surrounding objects. To further improve accuracy, the researchers integrated a convolutional neural network based on the single-shot multibox detector (SSD) algorithm.

These innovations allow the proposed automatic OWPT system to sequentially target multiple PV receivers of different sizes and at different distances and quickly switch between them without interruption. In experiments, the system worked seamlessly in both bright and dark environments, providing efficient and stable power transmission up to a distance of 5 meters.

“Our automatic OWPT system provides a stable and versatile wireless power transfer solution,” said Miyamoto. “This will play a key role in building sustainable IoT infrastructure, especially in smart factories, homes, and indoor environments where safe, wireless, dynamic, and scalable power supplies are essential.”